Dsd

HD

a

ARRAA

KTSHSV

1

oh1[sthat3rmt

agtadt

0d

Journal of Pharmaceutical and Biomedical Analysis 49 (2009) 950–956

Contents lists available at ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis

journa l homepage: www.e lsev ier .com/ locate / jpba

evelopment and validation of a reversed-phase HPLC method for theimultaneous analysis of simvastatin and tocotrienols in combinedosage forms

azem Ali, Sami Nazzal ∗

epartment of Basic Pharmaceutical Sciences, College of Pharmacy, University of Louisiana at Monroe, 1800 Bienville Drive, Monroe, LA 71201, United States

r t i c l e i n f o

rticle history:eceived 6 December 2008eceived in revised form 29 January 2009ccepted 10 February 2009

a b s t r a c t

A RP-HPLC method for the simultaneous analysis of tocotrienol isoforms (TRF) and simvastatin (SIM)in SIM–TRF nanoparticles (NPs) was developed. Analytes were monitored by UV detection at 238 and295 nm for SIM and TRF, respectively, using a gradient methanol/water elution. Calibration curves forTRF and SIM were linear over concentration range of 20–80 �g/mL and 1–10 �g/mL with correlation

vailable online 20 February 2009

eywords:ocotrienolimvastatinPLC

®

coefficients 0.9990 and 0.9991, respectively. The recovery of TRF and SIM from the NPs was in the rangefrom 97.35 to 102.19% and from 92.71 to 104.35%, respectively. This developed method was successfullyemployed in quantifying both drugs in NPs for future use in cancer therapy.

© 2009 Elsevier B.V. All rights reserved.

olid lipid nanoparticles (SLN )alidation

. Introduction

Statins, such as simvastatin (SIM, Fig. 1), represent a classf drugs that are commonly used for the management ofypercholesterolemia [1,2]. Chemically, SIM is 2,2-dimethyl-,2,3,7,8,8a-hexahydro-3,7-dimethyl-8-1naphthylenyl ester, [1S-1�,3�,7�,8�(2S*,4S*),-8a�]] butanoic acid. It is frequently pre-cribed for the treatment of hypercholesterolemia and was showno significantly decrease the mortality associated with coronaryeart disease [3,4]. SIM, however, is a pro-drug. After its oraldministration, SIM is hydrolyzed in the liver to its active form,he �-hydroxy acid, which competitively inhibits 3-hydroxyl--methylgutarylcoenzyme A (HMG-CoA) reductase. HMG-CoAeductase is an enzyme involved in the conversion of HMG-CoA toevalonate, which is the rate-limiting step in cholesterol biosyn-

hesis [2].Inhibition of the HMG-CoA reductase and subsequently block-

de of mevalonate synthesis was also shown to decrease tumorrowth in vivo. Therefore, statins were evaluated for their poten-

ial use in cancer therapy [5–8]. The antiproliferative and apoptoticctivity of SIM against breast cancer cells, for examples, wasemonstrated in recent studies [9–11]. Similarly, palm oil rich inocotrienols, commonly referred to as tocotrienol-rich-fraction or

∗ Corresponding author. Tel.: +1 318 342 1726; fax: +1 318 342 1737.E-mail address: nazzal@ulm.edu (S. Nazzal).

731-7085/$ – see front matter © 2009 Elsevier B.V. All rights reserved.oi:10.1016/j.jpba.2009.02.009

TRF (Fig. 2), was shown to display potent antiproliferative and apop-totic activity against breast cancer cells, which was also attributedto HMG-CoA reductase down-regulation [12–15]. TRF is an oilymixture of tocopherols and tocotrienols, in which tocotrienols con-stitutes 70–80% of the blend [12]. Both tocopherols and tocotrienolshave similar chemical structure characterized by a phytyl side chainattached to a chromane ring. The fundamental structural differencebetween the two groups is the phytyl chain, which is unsaturated intocotrienols (Fig. 2). Isoforms of tocopherols and tocotrienols differfrom each other by the degree of methylation of the chromane ring.

Due to their similar mechanism of activity, blends of statinsand TRF were evaluated for their therapeutic effects. It was foundthat combined low-dose treatment of �-tocotrienol or TRF and SIMdemonstrates synergistic antiproliferative effects against mam-mary tumor cells. These findings strongly suggested that a SIM–TRFcombination therapy could provide significant health benefits inthe prevention and/or treatment of breast cancer and would avoidthe myotoxic side effects caused by high-dose administration of SIMif taken alone [16].

Based on the observed synergistic effect, SIM–TRF nanoparti-cles (NPs) were manufactured for potential targeted therapy ofbreast cancer. This necessitated the development of a quantita-

tive and sensitive HPLC method that could discriminate betweenSIM, TRF isoforms, and the excipients used in the preparationof the NPs. Several HPLC methods were reported for the analy-sis of SIM alone [17,18] or in combination with other drugs, suchas ezetimibe [19,20]. Similarly, HPLC methods have been devel-

H. Ali, S. Nazzal / Journal of Pharmaceutical and

otmtosamwor

2

2

s2�iMta

k = (1)

Fig. 1. Chemical structure of simvastatin (SIM).

ped and validated for the analysis of individual tocotrienols andocopherols isoforms in human plasma [21]. None of the reported

ethods, however, addressed the analysis of composite TRF mix-ure of tocopherols and tocotrienols in the presence or absencef statins. The objective of this study was therefore to develop aimple, sensitive, and precise HPLC method for the simultaneousnalysis of TRF and SIM for use as a QC tool in the develop-ent of SIM–TRF NPs. The HPLC method reported in this studyas validated by assessing system suitability, specificity, limitf quantitation and detection, linearity, precision, accuracy, andecovery.

. Experimental

.1. Chemicals and reagents

Simvastatin was purchased from Haorui Pharma-Chem Inc. (Edi-on, NJ); tocotrienol-rich-fraction of palm oil (TRF, which contains0.2% �-tocopherol, 16.8% �-tocotrienol, 44.9% �-tocotrienol, 14.8%

-tocotrienol, and 3.2% of a non-vitamin E lipid soluble contam-

nants) was a gift from the Malaysian Palm Oil Board (Selangor,alaysia); Compritol 888 ATO US/NF (glyceryl behenate, a mix-

ure of ∼15% mono-, 50% di-, and 35% triglycerides of beheniccid [C22]) was provided by Gattefossé (Paramus, NJ), Lutrol® F

Fig. 2. Generalized chemical st

Biomedical Analysis 49 (2009) 950–956 951

68 NF (poloxamer 188) was obtained from BASF (Florham Park,NJ); methanol HPLC grade, chloroform, (±)-�-tocopherol (usedas internal standard, IS), and sodium hydroxide were purchasedfrom Sigma–Aldrich Co. (St. Louis, MO); alcohol USP (95%) waspurchased from AAPER Alcohol and Chemical Co. (Shelbyville, Ken-tucky); water was obtained from NanoPure purification system.All chemicals were used as received without further modifica-tion.

2.2. Instrumentation and chromatographic conditions

The HPLC analysis was carried out by a SpectraSystem appa-ratus (Thermo Electron Corporation, San Jose, CA) equippedwith a SCM1000 vacuum degasser, a P2000 pump, an AS1000-010 autosampler unit, and a UV3000-160 UV/Visible variablewavelength detector. A C18 (4.6 mm × 100 mm) Onyx® monolithicanalytical column (Phenomenex®, Inc., Torrance, CA) was usedfor the analysis. The column was supported with an analyti-cal guard cartridge system (Phenomenex® Inc., Torrance, CA).Data acquisition was performed using a chromatography softwareChromQuestTM version 4.2 (Thermo Electron Corporation, San Jose,CA). The column was flushed with methanol at a flow rate of3 mL/min for 5 min prior to the analysis of each set of three samples.For the separation and identification of TRF isoforms, an isocraticelution using 5% (v/v) water in methanol at a flow rate of 0.8 mL/minwas used. The mobile phase was allowed to run for 5 min prior toeach analysis. The detection of TRF was carried out at �max = 295 nm.For the simultaneous analysis of SIM and TRF, a gradient elution wasinitially carried out using a 15:85 (v/v) water/methanol system anda flow rate of 0.8 mL/min for 5 min. During this time, �max was setto 238 nm for the detection of SIM. Thereafter, �max was adjustedto 295 nm for the detection of TRF isoforms. The composition andflow rate of the mobile phase were also adjusted to a 5:95 (v/v)water/methanol blend and a flow rate of 1 mL/min. An injectionvolume of 20 �L was used in all experiments and the elution wascarried out at controlled room temperature (20–25 ◦C). For methoddevelopment and optimization, retention factor (k) was calculatedusing the following equation:

tR − tM

tM

where tR is the elution time for the most retained isoform of TRF,which corresponds to �-tocopherol; tM is the elution time of thesolvent front.

ructure of TRF isoforms.

9 al and

2

wtttfttscwwmstcccasrcl

2

11TtowaIrfiwl

2

(schtwuCftccbfUunspwu

52 H. Ali, S. Nazzal / Journal of Pharmaceutic

.3. Preparation of standard solutions

The stock solution of TRF was prepared by dissolving accuratelyeighed 10 mg in 10 mL of methanol to obtain a final concentra-

ion of 1.0 mg/mL. The prepared stock solution was stored at −10o −20 ◦C in amber glass scintillation vial. From this stock solu-ion, standards within a 20–80 �g/mL concentration range werereshly prepared prior to analysis. The stock solution of (±)-�-ocopherol, as the internal standard, was prepared by dissolvinghe IS in methanol at a final concentration of 1 mg/mL. The ISolution was stored under conditions similar to that for TRF. Aonstant concentration of 1 �g/mL of the IS [(±)-�-tocopherol]as used for all calibration points. Similarly, stock solution of SIMas prepared by dissolving accurately weighed 10 mg in 100 mL ofethanol to obtain a final concentration 100 �g/mL. The prepared

tock solution was stored at −10 to −20 ◦C in amber glass scintilla-ion vial. From this stock solution, standards within a 1–10 �g/mLoncentration range were freshly prepared prior to analysis. Tripli-ate 20-�L injections were made for each concentration of bothompounds and were analyzed under the conditions describedbove. The mean peak area ratio of TRF/IS and SIM/IS of eachtandard solution were then plotted as a function of their cor-esponding concentration in order to construct the calibrationurves and subsequently calculate the equations of the calibrationines.

.4. Construction of calibration curves for individual TRF isoforms

TRF calibration standards were prepared at concentrations of0, 20, 40, 60, 80, and 100 �g/mL from a standard solution of00 �g/mL by appropriate dilution with methanol. The individualRF isoforms (�-tocotrienol, �-tocotrienol, �-tocotrienol, and �-ocopherol) were identified by injecting onto the column a standardf known quantity of each isoform separately. Calibration curvesere then constructed by plotting the peak ratio of each isoform/IS

gainst TRF concentrations. JMP statistical software package (SASnstitute Inc., version 7.0) was subsequently used to perform linearegression analyses and test the fitted models. Correlation coef-cients were calculated and the results of the statistical analysesere considered significant if their corresponding p-values were

ess than 0.05.

.5. Preparation of SIM–TRF NPs

SIM–TRF NPs were manufactured using Lutrol® F68polyethylene–polypropylene block copolymer) as the primaryurfactant and Compritol® 888 ATO (glycerol behenate) as the lipidore encapsulating TRF and SIM. Hot O/W microemulsion with aigh-shear homogenization technique was used to manufacturehe SIM–TRF NPs [22]. Briefly, Compritol® 888 ATO, SIM and TRFere dissolved in chloroform, which was allowed to evaporatender a stream of nitrogen. The remaining lipid residue (containingompritol®, TRF and SIM) was melted at 80 ◦C. An aqueous sur-

actant solution containing 0.125% (w/v) of Lutrol® F68, preheatedo 80 ◦C, was then added to the molten lipid blend. The finaloncentration of the lipid blend in the solution was 0.25% (w/v),ontaining 1 mM of SIM and TRF each. The dispersion of the lipidlend in the surfactant solution was homogenized at 20,000 rpmor 5 min using IKA® Ultra-Turrax T8 mixer (IKA® Works Inc., NC,SA). The formed hot microemulsion was then sonicated for 10 minsing ultrasonic homogenizer (Biologics, Inc., VA, USA). SIM–TRF

anoparticle dispersions were subsequently formed by cooling theonicated microemulsion overnight in the refrigerator. The meanarticle size (±S.D.) of the SIM–TRF NPs was 253.7 nm (±0.45),hich was measured by photon correlation spectroscopy (PCS)sing a NicompTM 380 ZLS particle size analyzer (PSS Inc., Santa

Biomedical Analysis 49 (2009) 950–956

Barbara, CA). Prior to HPLC analysis, NP samples were preparedby diluting 100 �L of the SIM–TRF nanoparticle dispersion with900 �L of methanol.

2.6. Determination of SIM–TRF NPs entrapment efficiency

The entrapment efficiency of the SIM–TRF NPs was determinedby measuring the concentration of the free unloaded compoundsin the aqueous phase of the nanoparticle dispersion. Centrifugationwas carried out using Centrisart filters (molecular weight cutoff20,000 Da). Approximately 2.5 mL of the nanoparticle dispersionwas placed in the outer chamber of the filter assembly. The assem-bly was then centrifuged at 5000 rpm (approximately 3000 × g) for15 min at 5 ◦C using Eppendorf® 580R centrifuge (Hamburg, Ger-many). The NPs along with encapsulated compounds remainedin the outer chamber whereas the aqueous dispersion mediumcontaining the free unloaded compounds moved to the samplerecovery chamber through the filter membrane. After separation,the amount of the free compounds in the dispersion medium wasestimated by HPLC. The entrapment efficiency was subsequentlycalculated from the following equation:

entrapment efficiency (%)

= total amount of drug − amount of free drug in the aqueous dispersiontotal amount of drug

× 100

2.7. Method validation for simultaneous TRF/SIM assay

2.7.1. System suitabilityThe system suitability was evaluated by six replicate analyses

of a TRF/SIM mixture at a concentration of 60 �g/mL of TRF and20 �g/mL of SIM. The acceptance limit was ±2% for the percentcoefficient of variation (%CV) of the peak area and the retentiontime of TRF and SIM.

2.7.2. Detection and quantitation limits (sensitivity)Limits of detection (LOD) and limit of quantitation (LOQ) were

estimated from the signal-to-noise ratio [23,24]. LOD is defined asthe lowest concentration resulting in a peak area of three timesthe baseline noise. LOQ is defined as the lowest concentration thatprovides a signal-to-noise ratio higher than 10, with precision (%CV)and accuracy (%bias) within their acceptable range (10%).

2.7.3. Linearity (calibration curve)The calibration curves were constructed with eight concentra-

tions (simultaneously prepared) ranging from 1 to 10 �g/mL andfrom 20 to 80 �g/mL for SIM and TRF, respectively. Calibrationcurves were constructed by plotting the ratio of the mean peakarea of either �-tocotrienol or SIM to IS versus the concentration.The linearity was assessed by linear regression analysis, which wascalculated by the least square method.

2.7.4. Accuracy and precisionPrecision of the assay was determined by repeatability (intra-

day) and intermediate precision (inter-day) for 3 consecutive days[23,24]. Three different concentrations of SIM and TRF were ana-lyzed in six independent series in the same day (intra-day precision)and 3 consecutive days (inter-day precision). Every sample wasinjected in triplicate. The accuracy of the method, which is defined

as the nearness of the true value and found value [23,24], was eval-uated as %bias for TRF and SIM according to the following equation:

%accuracy = observed concentrationnominal concentration

× 100

H. Ali, S. Nazzal / Journal of Pharmaceutical and Biomedical Analysis 49 (2009) 950–956 953

Table 1Regression characteristics of the TRF isoforms determined by the HPLC method.

Analyte ANOVA Average retention time (tR) ± S.E.

Average peak areaa ± S.E. r2 b p-value Regression equationc

�-Tocotrienol 4.231 ± 0.2034 0.9338 <0.0001 y = 1.631 + 0.1691x 4.0533 ± 0.0622�-Tocotrienol 6.559 ± 0.0628 0.9990 <0.0001 y = 1.535 + 0.4617x 4.76 ± 0.0808�-Tocotrienol 3.897 ± 0.4376 0.9278 <0.0001 y = 0.7767 + 0.2357x 5.4366 ± 0.0949�-Tocopherol 19.833 ± 2.0204 0.2780 0.0245 y = 12.0 + 0.0953x 9.94 ± 0.1960

a Average peak area ×105.b R-squared.c y: estimated peak area (×105) at x TRF concentration (�g/mL).

Table 2System suitability study.

Parameters TRF (60 �g/mL) SIM (20 �g/mL) Internal standard (1 �g/mL)

Retention time (min) Peak area Retention time (min) Peak area Retention time (min) Peak area

MS%

2

abeapsFcibhtcy1p

2

Qta

Fa

ean (n = 6) 10.47 24433.12 3.93.D. 0.06 92.94 0.027CV 0.58 0.38 0.69

.7.5. SpecificityThe specificity of an analytical method may be defined as the

bility to detect the analyte peak in the presence of the analytey-products, or other inactive components, such as dosage formxcipients or impurities [23,24]. In this method, detection of TRFnd SIM in the presence of de-esterified SIM, as the major by-roduct of SIM degradation, as well as nanoparticle lipids andurfactants excipients was used as a measure of its specificity.orced degradation of SIM was carried out under basic hydrolyticonditions as described previously [25]. Briefly, SIM was dissolvedn a 70% ethanol solution containing 0.1 M NaOH and then incu-ated at 50 ◦C for 1 h to convert SIM to its corresponding open-ringydroxyl derivative. The de-esterified SIM and reference SIM solu-ions were then mixed into a blend containing 20 �g/mL of eachomponent. The blend was injected into the column for HPLC anal-sis using water/methanol system (15:85, v/v) at a flow rate ofmL/min with a �max adjusted to 238 nm. All experiments wereerformed in triplicates.

.7.6. StabilityThe stability of the drug solution was determined by analyzing

C samples after a short-term storage at controlled room tempera-ure (20–25 ◦C) for 12 h. The long-term stability was determined bynalyzing samples stored at 4 ◦C for 30 days. The autosampler sta-

ig. 3. TRF chromatogram at 0.8 mL/min isocratic elution using water/methanol atratio of 5:95 (v/v). �max was adjusted to 295 nm.

195714.67 12.95 7534.24883.52 0.037 22.45

0.45 0.29 0.30

bility was determined by analyzing the samples after 24 h of storagein the autosampler.

2.7.7. RecoveryThe absolute recovery was computed from the peak area

of TRF/SIM methanolic standard solutions to those containingTRF/SIM in the NPs at four different concentrations.

2.8. Statistical analysis

Data collected in this study were analyzed using JMP statisticalsoftware package by one-way analysis of variance (ANOVA). Uni-variate linear regression analysis using least square method wasapplied to test the fitted model. Correlation coefficient was cal-culated and the results of the statistical analysis were consideredsignificant if their corresponding p-values were less than 0.05.

3. Results and discussion

3.1. Method development and optimization

The chromatographic conditions were optimized for the simul-taneous determination of TRF and SIM within a short analysis time(<15 min) and an acceptable peak resolution (Rs > 2). To accomplishthese objectives, the chromatographic column was first chosenbased on peak shapes and resolution. In preliminary experiments,the sample retention time increased with an increase in columnlength. In order to avoid long run-times, a C8 column was initiallyused. This, however, resulted in peak overlap between SIM and �-tocotrienol, which is a component of TRF, with a consequent peakshape inconsistency during TRF elution. Therefore, a C18 mono-lithic column (5 �m, 100 mm × 4.6 mm i.d.), which permits the useof high flow rate with consequent low increase in back-pressure,was subsequently used to better resolve �-tocotrienol from SIMpeak, reduce elution time, and obtain sharp peaks for individualTRF isoforms.

With regard to the mobile phase, an HPLC method for SIM wasdescribed in the USP, in which phosphoric acid and acetonitrile(40:60, v/v) was used as the mobile phase and C18 column as the

stationary phase (USP 28 monograph). The use of salts in the mobilephase, however, can affect the lifespan of the column [26]. In orderto avoid this drawback, a mobile phase containing various ratios ofwater and methanol was initially used. Water, however, produced ahigh retention factor for TRF isoforms. At 25% water, retention factor

954 H. Ali, S. Nazzal / Journal of Pharmaceutical and

Table 3Results of regression analysis of TRF and SIM determined by HPLC.

Analyte Mean r2 ± S.E. Mean slope ± S.E.(n = 6)

Mean intercepta ± S.E.(n = 6)

S�

(wdhifTpkoAewwudd

3i

gTttstt

illustrated previously. Table 3 lists the outcome of the statistical

TI

N

TTTTSSSS

TS

N

TTTTSSSS

IM 0.9991 ± 0.4052 1.6346 ± 0.0132 3.0005 ± 0.8539-Tocotrienol 0.9990 ± 0.1197 0.4617 ± 0.0035 1.535 ± 0.2155

a Intercept is expressed in �g/mL ×105.

calculated from Eq. (1)) for the most retained peak (�-tocopherol)as k = 12 (i.e. tM = 26). In addition, TRF isoforms could not beistinguished due to peak overlaps. This is presumably due to theigh content of water in the mobile phase, which produced tail-

ng. SIM and TRF are poorly soluble in water whereas they arereely soluble in organic solvents, such as methanol and acetonitrile.herefore, an increase in the percentage of methanol in the mobilehase from 75 to 85% by volume decreased the retention factor, i.e., by approximately 50% (k = 6.3; tM = 14.6). Furthermore, peak res-lution between TRF isoforms was in the acceptable range (Rs > 2).further increase in methanol (95%), however, resulted in very fast

lution with poor SIM resolution from the solvent front (Rs < 2). Thisas observed when an isocratic elution using 5% water in methanolas carried out at a flow rate of 0.8 mL/min. This was resolved bysing a gradient elution as outlined in Section 2. The internal stan-ard [(±)-�-tocopherol] was used to eliminate the interferencesue to the impurities in TRF (constitutes 3.2% of TRF).

.2. Development of an HPLC method for the separation of TRFsoforms

A single, isocratic and selective reversed-phase liquid chromato-raphic method was developed for the separation of TRF isoforms.RF methanolic solutions (10 and 100 �g/mL) were injected under

he chromatographic conditions outlined in Section 2. The selec-ivity of the RP-HPLC method is illustrated in Fig. 3, which shownatisfactory peak separation. Table 1 lists the results of the statis-ical analysis and the average peak area and retention times forhe TRF isoforms. Retention time (tR) ranged from 4.05 min for �-

able 4ntra-day and inter-day precision and accuracy results of TRF and SIM (n = 6).

ominal concentration (�g/mL) Day 1

Meana %CV %Bias

RF (20) 20.535 2.990 2.675RF (40) 40.7733 1.488 1.933RF (60) 61.436 1.302 2.394RF (80) 80.72667 0.888 0.908IM (1.0) 0.991 3.574 -0.833IM (2.0) 2.071 3.668 3.583IM (6.0) 5.983 1.101 -0.278IM (10.0) 10.101 1.551 1.017

a Mean found concentration (�g/mL).

able 5hort-term, long-term, and auto-sampler stability for TRF and SIM combined solutions (n

ominal concentration (�g/mL) Short-term stability Lo

%Meana S.D. %CV %M

RF (20) 107.675 1.296 6.020 10RF (40) 104.017 0.757 1.818 10RF (60) 100.45 1.64 2.721 10RF (80) 100.908 2.418 2.995 10IM (1.0) 101.0 0.036 3.597 10IM (2.0) 101.583 0.048 2.379 10IM (6.0) 97.694 0.524 8.944 9IM (10.0) 101.933 0.549 5.393 10

a Expressed as percentage of nominal concentration.

Biomedical Analysis 49 (2009) 950–956

tocotrienol to 9.94 min for �-tocopherol, which indicates a rapidanalysis of TRF isoforms within a 12-min run-time. The p-values forthe four isoforms in TRF were <0.05, which suggests significant lin-ear model fit. The correlation coefficient (r) value was in the rangefrom 0.278 for �-tocopherol to 0.999 for �-tocotrienol. Therefore,�-tocotrienol was used as an indicative tracer during the simul-taneous analysis of TRF and SIM, as discussed in the subsequentsections. It is worth noting that the retention time of the synthetic(±)-�-tocopherol, which was used as the internal standard, wasapproximately 7.89 min as opposed to the 9.94 min observed withthe natural �-tocopherol. This difference in retention time betweenthe two compounds was verified by spiking the column with a pure�-tocopherol, which was previously extracted from natural TRF.

3.3. Method validation of the TRF/SIM simultaneous assay

3.3.1. System suitabilityThis step was accomplished under the conditions stated in Sec-

tion 2. The %CV of peak area and retention time for TRF, SIM, and ISwere within 2%. This indicates the suitability of the system to ana-lyze TRF and SIM simultaneously (Table 2). To evaluate the columnefficiency, the number of theoretical plates for TRF and SIM were4872 and 14,696, respectively. The retention factor was 4.82 and 1.0for TRF and SIM, respectively.

3.3.2. Detection and quantitation limits (sensitivity)The LOD for TRF and SIM was found to be 7.5 and 0.5 �g/mL,

respectively. The LOQ was 20 and 1 �g/mL for TRF and SIM, respec-tively.

3.3.3. Linearity (calibration curve)The linearity of the calibration curves for TRF and SIM was

calculated and constructed by least square regression method as

analysis. The correlation coefficient (r2) for the standard calibra-tion curves for TRF and SIM were 0.999 and 0.9991, respectively.This indicates linearity of the peak area ratio of TRF or SIM to IS inthe range of 20–80 �g/mL and 1–10 �g/mL, respectively.

Day 2 Day 3

Meana %CV %Bias Meana %CV %Bias

20.251 3.091 1.258 20.293 1.694 1.46740.643 2.234 1.608 40.175 2.016 0.43861.27 1.937 2.117 61.603 1.366 2.67280.893 1.753 1.117 80.726 1.723 0.908

1.003 2.795 0.333 1.015 2.972 1.52.046 1.793 2.333 2.036 2.701 1.8335.995 0.0582 -0.083 6.001 1.103 0.028

10.103 1.968 1.033 10.068 1.742 0.683

= 6).

ng-term stability Auto-sampler stability

eana S.D. %CV %Meana S.D. %CV

8.508 0.711 3.276 103.133 1.015 4.9211.933 1.739 4.265 102.937 2.338 5.6790.172 2.001 3.329 103.227 2.066 3.3351.741 3.430 4.214 103.154 3.343 4.0515.0 0.092 8.832 106.166 0.089 8.4331.583 0.048 2.379 108.5 0.213 9.8537.027 0.506 8.698 102.916 0.297 4.8212.933 0.517 5.027 102.783 00.359 3.501

H. Ali, S. Nazzal / Journal of Pharmaceutical and Biomedical Analysis 49 (2009) 950–956 955

Fto

3

aw(tca

3

i

The data for short-term, long-term, and the autosampler stabil-ity of the TRF and SIM solutions are given in Table 5. As shown,the %mean found concentration was within the acceptable limit(90–110%).

Table 6Absolute recovery of TRF and SIM from SIM–TRF nanoparticles (n = 6).

Nominal concentration(�g/mL)

Meana

Standard methanolicsolution

Methanolic solutionof nanoparticles

% Recovery

TRF (20) 19.67 20.1 102.186TRF (40) 39.50 37.67 95.367TRF (60) 61.98 60.34 97.353TRF (80) 78.90 76.98 97.566

ig. 4. The chromatograms of nanoparticle excipients (A), nanoparticles (before fil-ration) containing TRF (20 �g/mL) and SIM (20 �g/mL) (B), and standard solutionsf TRF (20 �g/mL) and SIM (20 �g/mL).

.3.4. Accuracy and precisionAccuracy and precision for the QC samples during the intra-

nd inter-day run are given in Table 4. As shown, the data wereithin the acceptance criteria (i.e. 5%). One-day intra-day accuracy

expressed as %bias) ranged from 0.908 to 2.675 and from −0.833o 1.017 for TRF and SIM, respectively. The calculated intra-day pre-ision (expressed as %CV) after 1 day ranged from 0.888 to 2.990

nd from 1.101 to 3.668 for TRF and SIM, respectively.

.3.5. SpecificityIn order to confirm the specificity of the method for TRF and SIM

n the presence of nanoparticle excipients, three methanolic solu-

Fig. 5. The chromatograms of SIM and its by-product (de-esterified form with anopen lactone ring), each existing at a concentration of 20 �g/mL. The mobile phaseconsisted of water/methanol at 15:85 (v/v) ratio. The flow rate was set at 0.8 mL/minand �max was adjusted to 238 nm.

tions of the dissolved SIM–TRF NPs were injected into the HPLC.As discussed above, the specificity of an HPLC method is the abil-ity to detect the analytes under research in the presence of otheringredients, such as the lipids and surfactants in the NPs. Peakidentification was performed under the experimental conditionsstated previously. Fig. 4 shows representative chromatograms ofthe SIM/TRF methanolic solution as well as those for the NPs in thepresence and absence of the SIM–TRF blend. As shown, the signalsdetected during the analysis correspond to the TRF and SIM only.Nanoparticle excipients did not exhibit any peaks, and therefore nointerferences were detected as indicated by the absence of signalsin the chromatograms.

Similarly, the specificity of the method in the presence SIM by-products was demonstrated by analyzing samples containing thereference and de-esterified SIM solutions as described in Section2. Opening of the lactone ring to yield a hydroxy-derivative wasreported as the only by-product of SIM degradation [25,27–30]. Thechromatogram of SIM and its de-esterified form are shown in Fig. 5.SIM exhibited a retention time (tR) of 3.00 ± 0.63 min whereas itscorresponding degradation product had a tR of 1.72 ± 0.33 min. Noother interfering signals were observed during the elution of SIMeven when �max was adjusted to 295 nm for the quantification ofTRF.

3.3.6. Stability

SIM (1.0) 0.96 0.89 92.708SIM (2.0) 1.08 1.07 99.074SIM (6.0) 6.07 5.95 98.023SIM (10.0) 9.87 10.3 104.356

a Mean found concentration.

9 al and

3n

osecaotcabstrtac

4

bqnhfa

R

[

[[

[[[[[

[27] A. Alvarez-Lueje, C. Valenzuela, J.A. Squella, L.J. Nunez-Vergara, J. AOAC Int. 88(2005) 1631–1636.

56 H. Ali, S. Nazzal / Journal of Pharmaceutic

.3.7. Recovery and application of the HPLC method in SIM–TRFanoparticle analysis

Entrapment efficiency and absolute recovery were measured inrder to demonstrate the applicability of the HPLC method for theimultaneous determination of SIM and TRF in NPs. To calculate thentrapment efficiency of SIM and TRF in the NPs, the free unloadedompounds were separated from the SIM–TRF NPs using Centris-rt filter assemblies as described in Section 2. The concentrationf the free unloaded SIM and TRF in the filtrate, as measured byhe HPLC method, was 0.011 and 0.025 mmol%, respectively, whichorresponds to an entrapment efficiency of 98.95% (±0.62) for SIMnd 97.5% (±0.35) for TRF. The absolute recovery was computedy comparing the peak area of the TRF/SIM methanolic standardolutions to those containing SIM–TRF NPs at different concentra-ions. The results of this experiment are given in Table 6. The %ecovery of SIM and TRF from the NPs was in the range from 92.71o 104.35 and 97.35 to 102.19, respectively. This indicates the suit-bility of the developed method in simultaneously quantifying theoncentration of both drugs in the NPs.

. Conclusion

The RP-HPLC method developed in this study was shown toe rapid, sensitive, and accurate in simultaneously detecting anduantifying TRF isoforms and SIM either alone or in the presence ofanoparticle excipients, such as lipids and surfactants. This methodas the advantage of being specific for both drugs without the need

or additional sample preparation, such as the extraction of thective constituents.

eferences

[1] J.P. Desager, Y. Horsmans, Clin. Pharmacokinet. 31 (1996) 348–371.[2] T. Kantola, K.T. Kivisto, P.J. Neuvonen, Clin. Pharmacol. Ther. 64 (1998) 177–182.

[[

[

Biomedical Analysis 49 (2009) 950–956

[3] W.F. Hoffman, A.W. Alberts, P.S. Anderson, R.L. Smith, A.K. William, J. Am. Med.Assoc. 29 (1996) 849–852.

[4] D.W. Kaufman, J.P. Kelly, L. Rosenberg, T.E. Anderson, A.A. Mitchell, JAMA 287(2002) 337–344.

[5] I. Kaneko, Y. Hazama-Shimada, A. Endo, Eur. J. Biochem. 87 (1978) 313–321.[6] A.J. Habenicht, J.A. Glomset, R. Ross, J. Biol. Chem. 255 (1980) 5134–5140.[7] V. Quesney-Huneeus, H.A. Galick, M.D. Siperstein, S.K. Erickson, T.A. Spencer,

J.A. Nelson, J. Biol. Chem. 258 (1983) 378–385.[8] K.P. Fairbanks, L.D. Witte, D.S. Goodman, J. Biol. Chem. 259 (1984) 1546–1551.[9] M.J. Campbell, L.J. Esserman, Y. Zhou, M. Shoemaker, M. Lobo, E. Borman, F.

Baehner, A.S. Kumar, K. Adduci, C. Marx, E.F. Petricoin, L.A. Liotta, M. Winters,S. Benz, C.C. Benz, Cancer Res. 66 (2006) 8707–8714.

[10] H. Seeger, D. Wallwiener, A.O. Mueck, Exp. Clin. Endocrienol. Diabetes 111(2003) 47–48.

[11] M.A. Shibata, Y. Ito, J. Morimoto, Y. Otsuki, Carcinogenesis 25 (2004) 1887–1898.12] B.S. McIntyre, K.P. Briski, A. Gapor, P.W. Sylvester, Proc. Soc. Exp. Biol. Med. 224

(2000) 292–301.[13] P.W. Sylvester, B.S. McIntyre, A. Gapor, K.P. Briski, Cell Prolif. 34 (2001) 347–357.[14] S. Shah, P.W. Sylvester, Exp. Biol. Med. (Maywood) 229 (2004) 745–755.[15] G.V. Samant, P.W. Sylvester, Cell Prolif. 39 (2006) 563–574.[16] A. Thibault, D. Samid, A.C. Tompkins, W.D. Figg, M.R. Cooper, R.J. Hohl, J. Trepel,

B. Liang, N. Patronas, D.J. Venzon, E. Reed, C.E. Myers, Clin. Cancer Res. 2 (1996)483–491.

[17] M.S. Arayne, N. Sultana, F. Hussain, S.A. Ali, J. Anal. Chem. 62 (2007) 536–541.[18] B.G. Chaudhari, N.M. Patel, P.B. Shah, Drug Form. Clin. Methods 90 (2007)

1242–1249.[19] B.G. Chaudhari, N.M. Patel, P.B. Shah, J. AOAC Int. 90 (2007) 1242–1249.20] N. Ozaltin, E. Ucakturk, Chromatographia 66 (2007) S87–S91.21] S.P. Yap, T. Julianto, J.W. Wong, K.H. Yuen, J. Chromatogr. B: Biomed. Sci. Appl.

735 (1999) 279–283.22] P. Ahlin, J. Kristl, J. Smid-Kober, Acta Pharm. 48 (1998) 257–267.23] J. Ermer, J. Pharm. Biomed. Anal. 24 (2001) 755–767.24] G.A. Shabir, J. Chromatogr. A 987 (2003) 57–66.25] V.B. Wali, P.W. Sylvester, Lipids 42 (2007) 1113–1123.26] M. Ashfaq, I. Ullahkhan, S.S. Qutab, S. Naeemrazzaq, J. Chil. Chem. Soc. 52 (2007)

1220–1223.

28] B.G. Chaudhari, J. AOAC Int. 90 (2007) 1242–1249.29] J.J. Zhao, I.H. Xie, A.Y.B.A. Roadcap, J.D. Rogers, J. Mass Spectrom. 35 (2000)

1133–1143.30] M. Jemal, Z. Ouyang, Rapid Commun. Mass Spectrom. 14 (2000) 1757–1765.